Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    INDEX
    Explanations

    number one

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    	step
    -0.07
     nightmare
    -0.07
     approaching
    -0.07
    tolower
    -0.07
     relics
    -0.07
     trailer
    -0.06
    istinguish
    -0.06
     Replica
    -0.06
     вход
    -0.06
    istes
    -0.06
    POSITIVE LOGITS
     Describe
    0.06
    _latitude
    0.06
     是否
    0.06
    izu
    0.06
    [df
    0.06
    (man
    0.06
    LK
    0.06
    elerinin
    0.06
    nw
    0.06
     Ye
    0.06
    Act Density 0.030%

    No Known Activations